Solved: Best practices for increasing/decreasing budget on set schedule?

Hi there,

Thanks for reaching out!

Happy to give you some initial thoughts and guidance on your situation. It's a common question, and it's good you're thinking about how to optimise your spend based on performance trends you're already seeing. A lot of people just set a daily budget and let it run, so you're already a step ahead by noticing that your cost per install (CPI) is better on the weekends.

Let's get into the weeds of it. I'll walk you through my thoughts on the learning phase, the best ways to actually schedule your budget, and then some broader strategies that will likely have an even bigger impact on your CPI than just budget timing alone. There's quite a bit to unpack here, but I'll try to keep it straightforward.

We'll need to look at the learning phase... but don't panic

First off, your main question about the learning phase. It’s a valid concern. Any significant edit to an ad set can potentially push it back into the learning phase. Meta says a 'significant edit' can be a change to budget, bidding, targeting, creative, or optimisation event. The reason it does this is because the algorithm needs to re-explore and learn how to best deliver your ads with the new parameters you've given it. During this phase, performance can be a bit unstable and your CPI might fluctuate more than usual as it gathers the 50-ish conversions it needs to stabilise.

However, in your specific case, I wouldn't worrey about it too much. Making regular, predictable budget adjustments like increasing spend on a Friday and decreasing it on a Monday is something the system can often handle without a major reset every single time. The algorithm is smarter than many give it credit for, and it gets used to these patterns. You might see the status flicker to 'Learning' for a short while, but it's unlikely to cause the kind of prolonged instability you'd see from, say, completely changing your target audience or ad creative.

The key thing is that you're not making massive, unpredictable changes. A 20-30% budget increase or decrease is usually fine. If you were doubling your budget and then halving it again every few days, that might cause more disruption. The impact is also less severe on ad sets that are already mature and have a lot of historical data. The system has a better baseline to work from, so small tweaks are less of a shock.

Tbh, the potential gain from allocating more budget to your peak performance days (weekends) will almost certainly outweigh the minor instability from the learning phase being briefly re-triggered. The goal of paid advertising isn't to keep the 'Learning Phase' status bar happy; it's to get the most efficient results possible. If that means making a change that the system needs a few hours to adapt to, so be it. It's a cost of doing business, and in this case, a very small and worthwhile one.

I'd say you have a few ways to schedule your budget

Okay, so we've established that changing the budget is the right move. Now, what's the best way to actually do it? You've got a couple of options, each with their own pros and cons.

1. Manual Changes:
This is the most straightforward method. You literally go into Ads Manager on Friday morning and increase the budget, and then go back in on Monday morning and decrease it. It's simple, gives you full control, and requires no complex setup. The downside is that it's manual. You have to remember to do it. If you're on holiday, or ill, or just busy, it might get missed, and you'll lose out on that optimisation. It’s reliable if you are disciplined, but it’s not very scalable.

2. Automated Rules:
This is probably the best-practice solution for what you want to acheive. Meta has a feature called Automated Rules that lets you create 'if-then' conditions for your campaigns, ad sets, or ads. You could set up a rule like:
-> "If Day is Friday at 8:00 AM, then increase daily budget by 25% on ad set X."
And another one:
-> "If Day is Monday at 8:00 AM, then decrease daily budget by 20% on ad set X."

This takes the manual work out of it. Once set up, it just runs in the background for you. It's precise and reliable. You can set notifications so you know when the rules have run. This is what I'd typically recommend for this kind of regular, scheduled adjustment. It's the 'set it and forget it' approach that frees you up to focus on more strategic things.

3. Lifetime Budgets & Dayparting:
This is a slightly different way of thinking about it. Instead of a daily budget, you can set a Lifetime Budget for your ad set for a specific period (e.g., a week or a month). When you use a lifetime budget, you unlock the 'Ad Scheduling' option, also known as dayparting. This lets you tell Meta to only run your ads on specific days of the week or at specific times of the day.

So, you could set it to run with a higher intensity on weekends. The system will automatically pace the spend to be higher on those days. You can also specify the user's timezone, which is useful if you target multiple countries. For instance, you could tell it to run ads from Friday 5 PM to Sunday 11 PM in the viewer's local timezone. The algorithm then manages the budget allocation for you to maximise results within those windows. This can be very effective, but it’s a bit less flexible than daily budgets. You're locked into the lifetime budget and date range, so it's harder to make quick adjustments if something changes.

For your situation, I'd probably start with Automated Rules. It gives you the automation without sacrificing the flexibility of a daily budget.

You probably should look beyond just the budget timing

Adjusting budget by the day of the week is a smart tactic. But it is just that—a tactic. To get a truly transformative improvement in your CPI and scale your app installs, you need to look at the whole strategy. Often, the reason performance is better on weekends is because your target audience simply has more free time to browse their phones, discover new apps, and go through the installation process. That's a user behaviour you can't change. What you *can* change are the fundamental inputs of your campaign: who you're targeting, what you're showing them, and what happens after they click.

I've managed a lot of app install campaigns, including one where we drove over 45,000 signups at under £2 per signup across Meta, TikTok, and Google. The big wins didn't come from just budget timing; they came from methodical testing and optimisation of the core campaign components.

Let's break down what you should be looking at.

Your Targeting Strategy
This is the single most important factor. Are you reaching the right people? Or are you just hitting a broad audience and hoping for the best? For an app install campaign, your audience structure should look something like a funnel.

-> Top of Funnel (ToFu) - Cold Audiences: These are people who've never heard of you. For a new account, this is where you start.

• Detailed Targeting: Don't just target broad interests like 'Mobile Games' if you have a strategy game. Get specific. Target interests like 'Civilization (game)', 'Age of Empires', competitor game titles, specific genres. Think about the magazines they read, the influencers they follow, the other apps they use. The more specific the interest, the higher the chance it contains your ideal user. A common mistake is targeting an interest like 'Apple Inc.' to find iPhone users. Everyone who has an iPhone has that interest; it tells you nothing. You need to find interests that seperate your ideal user from the general population.
• Lookalike Audiences (LALs): Once you have enough data (at least 100 installs, but ideally 1,000+), you can create Lookalikes. These are Meta's most powerful tool. But don't just create a Lookalike of 'all website visitors'. You need to create Lookalikes based on high-intent actions. The further down the funnel the source audience, the better the Lookalike will be. For an app, the priority would be:
  1. Lookalike of people who made an in-app purchase.
  2. Lookalike of people who started a trial.
  3. Lookalike of people who completed registration.
  4. Lookalike of people who installed the app.

-> Middle/Bottom of Funnel (MoFu/BoFu) - Warm Audiences (Retargeting): These are people who have shown interest but haven't installed or converted yet.

• Website Visitors: Anyone who visited your app's landing page but didn't install.
• Video Viewers: People who watched a significant portion (e.g., 50%+) of your video ads. They're clearly interested.
• Social Engagers: People who liked, commented on, or shared your Facebook/Instagram posts.

You should have seperate campaigns or ad sets targeting these different funnel stages. The creative and messaging should be different for each. A cold audience needs an introduction to your app, while a retargeting audience might just need a reminder or a special offer to get them over the line.

Your Creative Strategy
After targeting, your ad creative is the next biggest lever. Are you split-testing different formats and messages? From my experience running campaigns for various software and app clients, some things consistently work well:

• Video is King: For apps, video is almost always better than static images. It lets you show the app in action. A simple screen recording showing the core gameplay loop or the main user benefit can be incredibly effective.
• Test UGC-style ads: User-Generated Content style creative often out-performs slick, highly-produced ads. This could be as simple as someone filming themselves on their phone talking about why they love your app. It feels more authentic and trustworthy. We've seen this deliver huge results for SaaS clients, and the principle is identical for apps.
• Clear Call-to-Action (CTA): Your ad needs to tell people exactly what to do. "Install Now", "Play for Free", "Download Today". Make it unmissable.
• Focus on the "What's in it for me?": Don't just list features. Show the benefit. Instead of "Advanced level-up system", try "Become an unstoppable hero". Instead of "Detailed analytics", try "See exactly how you're improving".

You should always be testing. Run one ad set with at least 3-5 completely different creatives (e.g., a gameplay video, a UGC testimonial, a carousel showing key features). Let Meta's algorithm find the winner, then iterate on that winning concept.

You'll need to think about your expected costs

You mentioned your CPI is "much better" on weekends, but what does that mean in real numbers? It’s helpful to have some benchmarks to understand what a "good" CPI actually is. Of course, this varies wildly based on the app niche, the country you're targeting, and your optimisation goal.

Based on our general campaign data, here are some very rough ballpark figures for conversion-type objectives. Let's call an app install a 'signup' for this purpose.

*Note: These are broad estimates. Your actual results will vary.

Where does your CPI fall within these ranges? If your weekday CPI is £10 and your weekend CPI is £7, you're making a good optimisation, but you're still in the higher end of the range. This would suggest your fundamental problem isn't the timing, but more likely the targeting or creative, as we discussed. If your weekday CPI is £2 and your weekend CPI is £1, then you're already doing brilliantly, and scheduling your budget is a fine-tuning exercise.

I remember one client where we drove over 45,000 signups at under £2 cost per signup across Meta, TikTok, Apple, and Google Ads. I also recall another instance, where we reduced a client's Cost Per User Acquisition for their medical job matching SaaS from £100 to £7 by overhauling their targeting and creative on Meta and Google Ads. These results didn't happen by just tweaking the budget schedule. It came from a complete overhaul of their targeting, creative, and landing page experience. For apps, the in-app experience is your landing page. If users install but then immediately churn because the onboarding is confusing or the value isn't clear, your effective CPI (cost to get an *active* user) is much higher.

I've detailed my main recommendations for you below:

This is a lot to take in, I know. So here’s a summary of the main advice I have for you, laid out as an actionable plan you can start to implement.

I hope this detailed breakdown gives you a much clearer picture. Your initial instinct to adjust budgets based on daily performance is a good one, but it's really just the tip of the iceberg. The real, sustainable growth and cost reduction will come from applying a more structured, strategic approach to your targeting and creative.

Implementing all of this correctly—setting up the right funnels, constantly testing creative, analysing the data, and knowing when to scale or shift budget—takes a lot of time and expertise. It's a full-time job in itself. While the advice I've given you here is a solid foundation, having an expert partner to manage the process can often accelerate results significantly and avoid costly mistakes along the way.

If you'd like to chat further about how we could apply our experience to your app and help you scale your installs more effectively, we offer a free, no-obligation initial consultation. We could take a proper look at your account together and map out a more concrete growth strategy.

Regards,

Team @ Lukas Holschuh